Mass spectrometric methods for the analysis of protein structures are being improved, applied and tested in independent and collaborative research projects in proteomics. Projects in separation methods and data processing are in progress. In the past year, progress has been made on implementation of a computational environment for mass spectrometric proteomic data processing;isotope labeled fusion proteins as protein standards;quantification of identified and unidentified components in sets of complex liquid chromatography-high resolution mass spectrometry data;understanding post translational modification in bacterial ribosomes;and multi-institutional collaborative studies on standards in proteomics. The processing and mining of mass spectrometric proteomic data requires a system allowing investigators desktop computer access to terabytes of data;the ability to repeatedly probe their data with emerging software tools;and the ability to transfer this data to public archival repositories. A versatile system has been implemented in LNT and will be shared with other proteomics groups in NIH. We sought to determine if biosynthetic concatenated labeled peptides (concatemers) are equivalent to whole labeled proteins as internal standards for isotope dilution mass spectrometry. Concatemer internal standards are planned for accurate quantification of proteins as well as stoichiometric measurements of a large number of proteins in complexes. Mass spectrometry provides a platform for these measurements by using multiple reaction monitoring to follow specific transitions of peptides as they fragment;however, internal standards are only accurate if they faithfully mimic proteolytic properties of full-length proteins. We proposed using signature peptides plus 12 amino acids (6 amino- and 6 carboxy-terminal) that are selected through mass spectrometric screening as well as the public databases and literature. Synthetic genes for the extended selected sequences are fused with affinity tags and expressed by cloning a synthetic gene into an expression vector and labeled using 13C and 15N arginine and lysine amino acids. A human serum albumin (HSA) concatemer was tested because native HSA is readily available as well as 15N-labeled full-length HSA as a laboratory standard. HSA concatemer concentration was measured with respect to a chemically synthesized strep tag 10-mer peptide using a standard curve. Time, temperature, and enzyme studies have been performed and optimized. Three of the five peptides in the HSA concatemer accurately mimic the tryptic properties of native HSA. Two are nearly identical, but may require further parameter optimization. Human urine studies are planned to compare concatemer performance for a clinical chemistry assay. New algorithms that serve as the basis for the bioinformatics tools used in the analysis of liquid chromatographic - tandem high resolution mass spectrometry data resulted in several open source software tools becoming viable alternatives to proprietary software. Unbiased protein analysis has been required to assess differences between control and treatment groups in several of our projects. The determination of relative expression levels of peptides in large data sets with biological replicates has remained challenging. The size and complexity of the data generated by LNT has consistently pushed the envelope of all the existing software. Three of the more advanced and actively developed tools were tested against data from a large PSD dataset previously collected in our laboratory;this set featured 3 biological replicates in each of 3 treatment groups resulting in 288 individual liquid chromatographic tandem mass spectrometry runs. The three open-source software tools chosen for comparison were XCMS, MZmine, and Mass++ . These tools do not require the identification of peptides through database searches prior to quantification of chromatographic signals. Progenesis, a commercial proprietary program, and DBParser, an in-house program requiring peak identification prior to analysis, are being used for comparison. To test the robustness of the integration methods, a commercial standard protein mixture (Universal Proteomics Standard) with varying concentration and acquisition method (profile/centroided) will be analyzed with all the programs. We have made progress on functionally characterizing the post-translational modification (PTM) beta-methylthio-aspartic acid 88 of the Escherichia coli protein S12. Our data indicates a correlation between the presence of this PTM and the transcription of anaerobic genes belonging to the FNR regulon. Briefly, mass spectrometry analysis and affinity pull-down assays were used to identify two proteins RimO and YcaO that are specifically linked to interacting with and modifying S12. Gene knockouts for both proteins revealed a dramatic decrease in the modified form of S12 (the RimO knockout resulted in a complete absence of modified S12) and an overlapping decrease in transcription of genes belonging to the FNR regulon (determined by DNA expression microarray). Further investigation revealed that an absence of the PTM results in a substantial decrease in the abundance of the transcriptional factor FNR required for these genes to be expressed. It therefore seems likely that the S12 PTM affects the mRNA specific translation of FNR although the molecular details are unknown. Dr. Kowalak planned and participated in trans-institutional proteomics informatics and standards studies through a leadership role in the Association of Biomedical Resource Facilities. One study focused on the ability of proteomics laboratories to identify phosphopeptides and localize the site(s) of phosphorylation. Three goals of the study were: (1) evaluate the consistency of reporting phosphopeptide identifications and phosphosite localization across participant laboratories, (2) characterize the underlying reasons why result sets may differ, and (3) produce a benchmark phosphopeptide dataset, spectral library and analysis resource. The study design utilized common dataset and sequence library. Participants were tasked with utilizing the informatics tools of their choice and to report their results with a fixed identification confidence (1% FDR). A common template was provided for reporting results. A second study on standards began development of a phosphopeptide standard for proteomics with a focus on post-digestion analysis of the reference materials. Although there are numerous strategies available to carry out phosphopeptide analysis, determining the phosphorylation state of proteins in complex samples remains a formidable challenge. The primary goal of this study was to provide each participating laboratory with an opportunity to evaluate its capabilities and approaches with regard to detecting phosphopeptides and identifying sites of phosphorylation.